Exercise - Implement intelligent search with full-text, vector, and hybrid queries
In this exercise, you implement different search approaches in an Azure SQL Database. You create full-text indexes, run vector searches using stored embeddings, and combine both techniques with hybrid search using Reciprocal Rank Fusion (RRF). You then compare how each search approach handles the same query and observe the differences in results and performance.
Note
To complete this exercise, you need an Azure subscription and be approved for Azure OpenAI access. If you need Azure OpenAI access, apply at the Azure OpenAI limited access page.
Launch the exercise and follow the instructions.
